It has been recognized for some time that the selection of donors for blood transfusions according to the recognized rules for typing and cross matching of blood does not always assure transfusions without mild, moderate or even severe and fatal reactions, although the surgical procedure was beyond reproach. Such unexplained reactions are frequent, much more frequent than the reports in the literature would indicate. They brought about an attitude of defeatism in surgeons who take it for granted that some reactions cannot be avoided and who accept that state of affairs as a necessary evil. This is particularly true with regard to reactions that are of only slight or moderate severity. It is not feasible to review the very large number of contributions dealing with transfusion reactions. Only the more recent ones and only those with sufficient data concerning the blood groups will be referred to. Recently de Gowin1